Significant consequences associated with cannabis abuse have been identified, and rates of enrollment into treatment for adolescent substance abuse with cannabis as the primary substance continue to rise. Family- based and individual interventions show efficacy, yet, rates of sustained abstinence are low, and the majority fails to reduce substance use substantially. This renewal application proposes to extend our focus on developing efficacious treatments for teen substance use disorders. An initial Stage I project showed that an intervention comprising behavioral therapy (BT), contingency management via clinic- and parent-delivered abstinence incentives (CM), and behavioral parent training (BPT) improved abstinence rates during and post treatment. Findings from a Stage 2 trial indicate that (1) the intervention has a positive effect on abstinence, and (2) BPT does not enhance outcomes above that achieved with BT/CM. Although both studies indicate enhanced outcomes, a considerable proportion of teens did not achieve a substantial period of abstinence and relapse was common. Importantly, we observed that if teens achieved abstinence, this almost always occurred by Week 6. The goal of this proposal is to further develop this intervention to increase rates of abstinence and decrease relapse rates. We will test two novel strategies to enhance outcomes. Working Memory Training (WMT), an efficacious method for strengthening specific cognitive processes, aims to improve factors (e.g., delay discounting / impulsive decision-making) that have shown a strong relation to substance use and treatment response. Second, more intensive and higher magnitude CM (ICM) will be used to motivate abstinence among teens who are not abstinent by Week 6. We hypothesize that these strategies will improve outcomes by modifying a fundamental cognitive system involved in making choices to engage in risky behavior and by increasing motivation to abstain in early nonresponders. Aim 1 will pilot and refine the new procedures in a community clinic in preparation for the randomized trial. A sequential, multiple assignment randomized trial (SMART) will allow us to determine the most effective first-line treatment and the most effective adaptive strategy (Aim 2). All teens wil begin treatment with BT/CM or BT/CM/WMT. After 6 weeks, responders will continue in their first-line treatments, while nonresponders will be randomized to ICM or to continue with first- lin treatment. Aim 3 will conduct mechanistic analyses to assess whether cognitive changes related to WMT engender increased abstinence, and whether specific tailoring variables moderate treatment effects. Aim 4 will gather formative data on implementation factors to inform future large-scale studies and dissemination efforts. Primary hypotheses are: (1) first-line treatment with WMT will improve abstinence outcomes and reduce relapse; (2) strategies with ICM for nonresponders will result in better outcomes than those without; (3) WMT will reduce delay discounting, which will predict outcome. The unique approach holds promise for reducing multiple types of risky behaviors by affecting basic mechanisms that determine impulsive decision-making.